Aspheric mold parameter optimization design method and system based on dynamic modeling

Abstract

The application relates to the technical field of aspheric mold parameter optimization, in particular to an aspheric mold parameter optimization design method and system based on dynamic modeling, which comprises the following steps: acquiring geometric data of a mold cavity; analyzing the geometric data to identify regions with key influence on the precision of a final product and regions with non-key influence; for the regions with key influence, deploying computing resources and performing simulation capable of reflecting microscopic physical processes; for the regions with non-key influence, performing simulation mainly considering computing efficiency; and exchanging and correlating physical field information between the simulation of the regions with key influence and the simulation of the regions with non-key influence to ensure the coherence of the overall simulation process. The application can improve the precision and reliability of the final product.

Description

Technical Field

[0001] This invention relates to the technical field of aspherical mold parameter optimization, specifically to a method and system for aspherical mold parameter optimization design based on dynamic modeling. Background Technology

[0002] In the field of precision manufacturing, the design and fabrication of aspherical molds are crucial in determining the final performance of optical components or precision mechanical parts. Especially in the manufacture of ultra-precision optical components where surface profile accuracy requirements reach the nanometer or even sub-nanometer level, traditional mold design methods often rely on engineers' experience and repeated trial and error. This is not only time-consuming and labor-intensive but also struggles to handle complex geometries, nonlinear material behavior, and microscopic deviations during processing. To overcome these bottlenecks, the industry has introduced mold parameter optimization design systems based on dynamic modeling. These systems aim to automatically find the optimal combination of mold geometry and process parameters by simulating the physical phenomena during injection molding.

[0003] However, in practical applications, especially when faced with the combined effects of various factors such as extremely complex geometric features, non-ideal response of special mold materials, micro-defects introduced by ultra-precision machining, non-uniformity of cooling conditions inside the mold, and viscoelastic deformation of materials during long-term service of the molded parts, the prediction and compensation capabilities of existing dynamic modeling methods are still insufficient, making it difficult for the final product to continuously and stably achieve the required limit accuracy and long-term reliability. Summary of the Invention

[0004] The purpose of this invention is to address the aforementioned shortcomings by proposing a method and system for optimizing the parameters of aspherical molds based on dynamic modeling.

[0005] The present invention adopts the following technical solution: A parameter optimization design method for aspherical molds based on dynamic modeling, comprising the following steps: Obtain the geometric data of the mold cavity; Analyze the geometric data to identify areas that have a critical impact on the accuracy of the final product, as well as areas that have a non-critical impact. For regions with critical impacts, computing resources are allocated, and simulations that reflect microscopic physical processes are conducted. For areas with non-critical impact, simulations are conducted with computational efficiency as the primary consideration. The simulation of critically affected areas is conducted by exchanging and correlating physical field information between the simulation of non-critically affected areas to ensure the coherence of the overall simulation process. The physical field information includes data on temperature field, pressure field, flow velocity field, and stress field.

[0006] This technical solution enables the optimized design of aspherical mold parameters. By distinguishing between critical and non-critical influence areas and employing different simulation strategies, it effectively balances simulation accuracy and computational efficiency. Furthermore, through the exchange and correlation of physical field information, it ensures the continuity of the overall simulation process. This solves the problem of insufficient prediction and compensation capabilities in existing technologies when dealing with complex mold designs, thereby improving the accuracy and reliability of the final product.

[0007] This application also discloses a parameter optimization design system for aspherical molds based on dynamic modeling. Using the above method, the system includes: The module acquires the geometric data of the mold cavity; The identification module analyzes geometric data to identify areas that have a critical impact on the accuracy of the final product, as well as areas that do not have a critical impact. The critical region simulation module allocates computing resources to key affected areas and performs simulations that reflect microscopic physical processes. The non-critical area simulation module performs simulations for areas with non-critical impact, with computational efficiency as the primary consideration. The information exchange module facilitates the exchange and correlation of physical field information between the simulation of critically affected areas and the simulation of non-critically affected areas, ensuring the continuity of the overall simulation process. The physical field information includes data on temperature, pressure, flow velocity, and stress fields.

[0008] This technical solution provides a system that integrates data acquisition, region identification, hierarchical simulation, and information exchange, enabling the effective implementation of the parameter optimization design method for aspherical molds, thereby improving design efficiency and the accuracy of the final product.

[0009] This application significantly improves the accuracy and reliability of the parameter optimization design of aspherical molds, enabling the final product to consistently and stably meet the surface accuracy requirements at the nanometer or even sub-nanometer level, overcoming the limitations of traditional methods that are time-consuming, labor-intensive, and unable to cope with complex situations.

[0010] To further understand the features and technical content of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are for reference and illustration only and are not intended to limit the present invention. Attached Figure Description

[0011] Figure 1 This is a flowchart of a method for optimizing the parameters of an aspherical mold based on dynamic modeling, according to the present invention. Figure 2 This is a schematic diagram of the structure of an aspherical mold parameter optimization design system based on dynamic modeling according to the present invention. Detailed Implementation

[0012] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the spirit of the present invention. Furthermore, the accompanying drawings of the present invention are for simple illustrative purposes only and are not depictions of actual dimensions; this is stated in advance. The following embodiments will further describe the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention.

[0013] This embodiment provides a method and system for optimizing the parameters of aspherical molds based on dynamic modeling, combined with... Figure 1 and Figure 2 As shown.

[0014] refer to Figure 1 A parameter optimization design method for aspherical molds based on dynamic modeling, comprising the following steps: Obtain the geometric data of the mold cavity; Analyze the geometric data to identify areas that have a critical impact on the accuracy of the final product, as well as areas that have a non-critical impact. For regions with critical impacts, computing resources are allocated, and simulations that reflect microscopic physical processes are conducted. For areas with non-critical impact, simulations are conducted with computational efficiency as the primary consideration. The simulation of critically affected areas is conducted by exchanging and correlating physical field information between the simulation of non-critically affected areas to ensure the coherence of the overall simulation process. The physical field information includes data on temperature field, pressure field, flow velocity field, and stress field.

[0015] Among them, "geometric data of the mold cavity" refers to the digital description of the space inside the mold used to form the product shape, which usually includes information such as three-dimensional coordinates, surface equations, and topological structure. This data is the basis for subsequent simulation and analysis.

[0016] "Critical influence areas" refer to localized regions within the mold cavity that have a decisive impact on the final product's accuracy (especially surface profile accuracy). These areas may have complex geometric features, high curvature variations, or be subjected to extreme physical conditions during the molding process.

[0017] "Non-critical areas" refer to areas in the mold cavity that have a relatively small impact on the precision of the final product, or areas with simple geometric features and drastic changes in physical conditions.

[0018] "Simulation of microscopic physical processes" refers to high-precision, high-resolution numerical simulations of physical phenomena such as material flow, heat transfer, solidification, stress, and strain at the micrometer or even nanometer scale. This type of simulation can capture details that are difficult to detect in traditional macroscopic simulations, such as microscale eddies, cavitation phenomena, and local phase transitions.

[0019] "Physical field information" refers to various parameters describing the physical state inside the mold cavity during the simulation process, including the temperature field (temperature distribution at each point), pressure field (pressure distribution at each point), flow velocity field (fluid velocity distribution at each point), and stress field (stress distribution at each point). This information is exchanged between different simulation regions to ensure the physical consistency and coherence of the entire simulation process.

[0020] This application provides a method for optimizing the parameters of aspherical molds based on dynamic modeling, the specific implementation of which is as follows: First, the geometric data of the mold cavity needs to be obtained. This step can be achieved in several ways. For example, a 3D scanner can be used to perform a high-precision scan of the existing mold cavity to directly obtain its point cloud data, and then a precise geometric model can be reconstructed using reverse engineering software. Alternatively, if the mold is designed based on CAD design software, the geometric data of the mold cavity can be directly exported from the CAD file, such as STEP or IGES formats. Furthermore, the geometric model of the mold cavity can also be manually constructed in 3D modeling software based on the design drawings or actual mold dimensions through manual measurement and modeling.

[0021] Secondly, the acquired geometric data is analyzed to identify regions with critical impact on the final product's accuracy, as well as non-critical regions. This analysis can be achieved using geometric feature recognition algorithms. For example, geometric parameters such as curvature and gradient change rate at various points on the mold cavity surface can be calculated, identifying regions with drastic curvature changes, high gradient values, or complex freeform surfaces as critical impact regions. Conversely, regions with gentle curvature changes, low gradient values, or simple geometric shapes are identified as non-critical impact regions. Another approach is based on an empirical knowledge base. Typical key region features of different types of aspherical molds are pre-stored, and then the geometric data of the current mold is matched with the knowledge base to identify critical impact regions.

[0022] Next, computational resources are allocated to key affected areas, and simulations reflecting microscopic physical processes are performed. During implementation, more CPU cores, GPU accelerators, and memory resources from high-performance computing clusters can be allocated to these critical areas to support high-resolution mesh generation and complex physical model calculations. For example, molecular dynamics simulations, lattice Boltzmann methods, or higher-order finite element methods can be used to perform refined simulations of the flow, heat transfer, and solidification processes of melts within microstructures, capturing nonlinear behavior at the microscale.

[0023] Simultaneously, for regions with non-critical impact, simulations are conducted primarily based on computational efficiency. For these regions, simplified physical models and coarser meshes can be used to reduce computational load and time. For example, traditional macroscopic finite element analysis or the finite volume method can be employed, using lower-order element types. In terms of computational resource allocation, fewer CPU cores and memory can be allocated, and even load balancing strategies from parallel computing can be employed to distribute computational tasks for these regions to idle computing nodes, thereby maximizing overall computational efficiency.

[0024] Finally, physical field information is exchanged and correlated between the simulations of critically affected regions and non-critically affected regions to ensure the consistency of the overall simulation process. Physical field information includes data on temperature, pressure, flow velocity, and stress fields. Specifically, a virtual boundary can be set at the interface between the two regions. High-precision simulation results from the critically affected region (such as temperature, pressure, flow velocity, and stress distribution at the interface) can be transferred to the simulation of the non-critically affected region as boundary conditions using interpolation algorithms. Conversely, simulation results from the non-critically affected region can also serve as macroscopic constraints for the critically affected region. This information exchange can be unidirectional or bidirectional iterative to ensure the continuity and consistency of the physical fields at the interface between the two regions. For example, domain decomposition or multi-scale coupling methods can be used to iteratively solve the problem until the physical quantities at the interface converge, thereby achieving the consistency of the overall simulation process.

[0025] This application's method for optimizing aspherical mold parameters based on dynamic modeling refines the geometric data of the mold cavity and divides it into critical and non-critical influence regions according to their impact on product accuracy. For the critical influence regions, this application allocates more computational resources to perform high-precision simulations that reflect microscopic physical processes, thereby capturing microscale physical phenomena that are difficult to detect using traditional macroscopic simulations, such as micro-fluid dynamics, local thermal gradients, and subtle changes during the curing process. These microscopic phenomena have a decisive impact on the surface contour accuracy of the final product. Meanwhile, for the non-critical influence regions, this application prioritizes computational efficiency, employing a relatively simplified simulation method to avoid unnecessary computational overhead.

[0026] More importantly, this application establishes a mechanism for exchanging and correlating physical field information between the simulation of critical and non-critical influence areas. By transferring and coupling physical field data such as temperature, pressure, flow velocity, and stress at the regional boundaries, the continuity and physical consistency of the entire simulation process are ensured. This regional, multi-scale simulation strategy enables this application to balance simulation accuracy and efficiency, accurately predicting the microscopic deformation of critical areas while efficiently optimizing the parameters of the overall mold.

[0027] This application significantly improves simulation efficiency by introducing a distinction between critical and non-critical influence regions and selectively allocating computational resources for simulations of varying precision. For example, when dealing with precision optical component molds with extremely complex geometries, this application can concentrate computational resources on the local areas that have the greatest impact on the final product's precision, performing high-resolution simulations of microscopic physical processes. This accurately captures the nonlinear thermal expansion / contraction characteristics of the mold material under rapid thermal cycling, the microstructural changes induced by ultra-precision machining, and the sub-nanometer-level surface morphology errors and local residual stresses generated during mold manufacturing. Meanwhile, for other non-critical regions, efficient macroscopic simulations are employed, avoiding unnecessary computational burdens.

[0028] Furthermore, this application ensures the continuity of the overall simulation process by exchanging and correlating physical field information between different simulation regions. This effectively solves problems such as the differentiated curing behavior of polymers caused by uneven local cooling conditions inside the mold cavity, and the slow drift of surface contours caused by polymer viscoelastic relaxation due to residual stress inside the molded part under long-term environmental effects. This coupled simulation method enables this application to more comprehensively and accurately predict and compensate for microscopic deformations that originate from multiple sources, interact with each other, and change over time. This allows the mold design scheme generated by the optimized system to consistently and stably achieve nanometer-level or even sub-nanometer-level surface accuracy requirements in actual production, especially in critical local areas and under long-term service conditions. Therefore, this application represents a significant technological advancement and innovation in improving the accuracy, efficiency, and reliability of aspherical mold parameter optimization design.

[0029] This application further proposes steps for allocating computational resources and conducting simulations that reflect microscopic physical processes, targeting regions with key impacts: In the critical areas of the mold cavity, deploy a micro-sensor array to collect local temperature, pressure, and flow rate data; Analyze local temperature, pressure and flow rate data to identify transient microscale deformation sensitive points. These microscale deformation sensitive points refer to points in the mold cavity that cause deformation due to micro-fluid dynamics, local thermal gradients or changes during the curing process. For microscale deformation-sensitive points, a local high-resolution solver is activated to simulate the local flow, transient heat transfer, and polymer solidification process of the melt under the influence of microscale eddies or cavitation, and the simulation results are obtained. Based on the simulation results, the deformation contribution of microscale deformation-sensitive points to the final product surface profile is quantified. A compensation strategy is generated based on the deformation contribution.

[0030] Specifically, micro-sensor arrays can include thermocouples, piezoresistive sensors, and micro-Pitto units, deployed inside the mold cavity, particularly in areas with complex geometry, dramatic curvature changes, or anticipated stress concentrations. By acquiring local temperature, pressure, and flow rate data in real-time and at high frequency, accurate transient physical field information can be provided for subsequent microscale deformation analysis. Analysis of the acquired local temperature, pressure, and flow rate data aims to identify transient signals that significantly deviate from normal steady-state or macroscopic trends using data processing algorithms (such as wavelet analysis, Fourier transform, or machine learning models). These transient signals often indicate microscopic hydrodynamic phenomena (such as local eddies or shear thinning), local thermal gradients (such as rapid cooling or non-uniform heat transfer), or phase transition inhomogeneities during polymer curing. Microscale deformation-sensitive points can be understood as localized areas within the mold cavity where these microscopic phenomena induce nanoscale deformation of the final product's surface contour. In practical applications, once a microscale deformation-sensitive point is identified, a specially designed local high-resolution solver is activated. The solver is configured to perform fine mesh generation and high-precision calculations within these specific regions to simulate the complex behavior of melts at the microscale. For example, it can simulate eddies formed in the melt at tiny gaps or corners, cavitation caused by sudden pressure drops, and how these phenomena affect the transient heat transfer efficiency and polymer solidification kinetics in local areas. This locally refined simulation provides detailed results regarding the material behavior at deformation-sensitive points. Furthermore, based on the simulation results obtained from this local high-resolution solver, techniques such as numerical integration, finite element analysis, or boundary element method can be used to map the local deformations (e.g., thermal expansion / contraction, viscoelastic deformation, solidification shrinkage) at microscale deformation-sensitive points onto the surface profile of the final product, thereby quantifying their specific deformation contribution to the accuracy of the product's surface profile. Based on this quantified deformation contribution, corresponding compensation strategies can be generated. This strategy may include fine-tuning the geometry of the mold cavity, local optimization of injection process parameters (such as injection speed, holding pressure, and cooling time), or auxiliary correction measures introduced in subsequent processing to offset or minimize deformation caused by microscale deformation-sensitive points.

[0031] This application's solution utilizes a micro-sensor array deployed in key areas of the mold cavity to achieve real-time, high-precision acquisition of local temperature, pressure, and flow rate data. This enables the system to capture transient microscale physical phenomena that are difficult to detect in traditional macroscopic simulations. It is precisely because of this acquisition of detailed data that the system can accurately analyze and identify transiently emerging microscale deformation-sensitive points, thus focusing attention on the local areas that have the greatest impact on the final product's accuracy. Subsequently, a local high-resolution solver is activated for these sensitive points, enabling precise simulation of the complex flow, heat transfer, and solidification processes of the melt at the microscale, revealing the specific mechanisms by which these processes affect the product's surface contour deformation. By quantifying the deformation contribution caused by these microscale deformation-sensitive points, this application provides a precise basis for generating subsequent compensation strategies, effectively solving the problem of insufficient accuracy in handling complex microscale deformations using traditional methods.

[0032] Through the above technical solution, this application can accurately identify, quantify, and compensate for microscale deformation phenomena in key influencing areas of aspherical mold cavities. Compared to basic solutions that only simulate general microscopic physical processes, this application introduces a micro-sensor array for local data acquisition, identifies transient microscale deformation sensitive points, and activates a local high-resolution solver for fine simulation, greatly improving the depth of understanding and prediction accuracy of the complex physical processes inside the mold cavity. This allows for a more accurate quantification of the deformation contribution of the final product's surface contour, thereby generating more refined and effective compensation strategies, ultimately significantly improving the manufacturing precision and consistency of aspherical products, especially under nanometer-level precision requirements, where its advantages are even more prominent.

[0033] In some preferred embodiments, it is assumed that a high-precision injection mold for an aspherical lens needs to be designed. First, by analyzing the geometric data of the mold cavity, the lens edge and center regions are identified as critical influence areas because the curvature changes drastically in these areas, significantly affecting optical performance. In these critical influence areas, an array consisting of miniature thermocouples, miniature piezoresistive sensors, and miniature flow rate sensors is deployed. During injection molding, these sensors collect local temperature, pressure, and melt flow rate data in real time. For example, when the melt flows through the tiny chamfer of the lens edge, the sensor data may show transient temperature drops and pressure fluctuations. Through data analysis, microscale eddies and microscale deformation sensitive points caused by localized rapid cooling are identified in this region. For these sensitive points, the system automatically initiates a local high-resolution solver based on the coupling of computational fluid dynamics and finite element analysis to perform detailed simulations of melt flow, transient heat transfer, and polymer solidification processes in this region. The simulation results show that these microscale eddies lead to uneven local material shrinkage, creating a local depression of approximately 50 nanometers on the surface profile of the final product. Based on this deformation contribution, the system generates compensation strategies, such as making micron-level geometric corrections to the cavity in this area during the mold design stage, or adjusting the local cooling water channels in the injection molding process to optimize heat transfer uniformity, thereby effectively offsetting or minimizing this microscale deformation and ensuring that the surface accuracy of the final lens meets the design requirements.

[0034] This application further proposes the following steps for analyzing local temperature, pressure, and flow velocity data to identify transiently emerging microscale deformation-sensitive points: Local temperature, pressure, and flow rate data are decomposed to separate signal components caused by different microscale physical phenomena; Based on the decomposed signal components, identify transient microscale deformation sensitive points.

[0035] Specifically, decomposing local temperature, pressure, and flow velocity data involves using signal processing techniques, such as wavelet decomposition, Fourier transform, independent component analysis, or empirical mode decomposition, to break down the raw, mixed sensor data into multiple independent or quasi-independent signal components. These signal components correspond to different microscale physical phenomena. For example, one component might primarily reflect the local flow characteristics of the melt under the influence of microscale eddies, another might reflect the transient heat transfer effect caused by a local thermal gradient, and still others might be related to volume shrinkage or phase transition during polymer solidification. This decomposition process effectively removes noise and interference components from complex signals and highlights physical information directly related to deformation-sensitive points. Identifying transiently emerging microscale deformation-sensitive points based on the decomposed signal components can be understood as applying corresponding physical models or threshold judgment rules to each or each group of specific signal components after separating the signal components caused by different physical phenomena. For example, for signal components reflecting local flow characteristics, the presence of abnormal eddies or cavitation phenomena can be determined by analyzing their amplitude, frequency, or phase changes; for signal components reflecting thermal gradients or solidification processes, their rate of change or local peak values ​​can be monitored to identify regions that may lead to uneven material shrinkage or expansion. This identification based on signal components of specific physical phenomena allows for more precise location and characterization of microscale deformation-sensitive points, avoiding the confusion and errors that may arise from directly identifying from mixed signals.

[0036] The proposed solution decouples complex original sensor data into multiple independent signal components by decomposing local temperature, pressure, and flow velocity data. Because these signal components correspond to different microscale physical phenomena, such as microscale eddies, local thermal gradients, or changes during polymer curing, it becomes possible to analyze each specific physical phenomenon individually. This approach avoids interference between signals from different physical phenomena, allowing for a clearer identification of the microscale physical mechanisms that truly cause transient deformation. This refined signal processing method ensures that the identification of microscale deformation-sensitive points is no longer based on fuzzy composite signals but on explicit physical causes, significantly improving the accuracy and reliability of identification and providing more precise input for subsequent simulation using a local high-resolution solver.

[0037] In some preferred embodiments, it is assumed that in the critical influence region of the aspherical mold cavity, a micro-sensor array collects temperature, pressure, and flow rate data encompassing multiple effects such as melt flow, local heat conduction, and polymer curing shrinkage. To accurately identify microscale deformation-sensitive points, these raw data are first processed by wavelet decomposition. Wavelet decomposition decomposes the raw signal into sub-band signals at different frequency scales. High-frequency components may correspond to transient microscale eddies or cavitation phenomena, while low-frequency components may reflect slow changes in local thermal gradients or the overall shrinkage trend during polymer curing. After separating these signal components, a threshold can be set for the high-frequency components. When the amplitude or rate of change exceeds this threshold, it is determined that there are deformation-sensitive points caused by microscale eddies or cavitation. For the low-frequency components, the deformation-sensitive regions caused by local thermal inhomogeneity or uneven curing can be identified by combining the material's thermal expansion coefficient and shrinkage rate model. This layered and refined analysis allows for more accurate location and characterization of microscale deformation-sensitive points. For example, it can identify minute depressions caused by local shear heat effects during the initial injection molding stage due to high-speed filling, or local warping caused by uneven curing during the cooling stage.

[0038] This application further proposes steps for quantifying the deformation contribution of microscale deformation-sensitive points to the surface profile of the final product based on simulation results, including: To obtain data on the microstructure evolution of mold materials under long-term thermal cycling; To obtain viscoelastic deformation data of polymer materials during long-term service; The data on the microstructure evolution of mold materials under long-term thermal cycling and the viscoelastic deformation data of polymer materials during long-term service are integrated with the deformation contribution of microscale deformation-sensitive points to the surface profile of the final product within a single injection cycle to obtain integrated data. Based on the integrated data, calculate the changes in the local thermal expansion coefficient and elastic modulus of the mold material; The changes in the local thermal expansion coefficient and elastic modulus of the mold material are introduced as correction terms into the calculation of the deformation contribution of microscale deformation-sensitive points to the surface profile of the final product within a single injection cycle. Based on the integrated data, the changing trend of local shrinkage of polymer materials is predicted, and the prediction results are obtained. By superimposing the correction terms and prediction results, the cumulative deformation contribution of microscale deformation-sensitive points to the final product surface profile throughout the entire product lifecycle is obtained.

[0039] Specifically, obtaining microstructural evolution data of mold materials under long-term thermal cycling refers to acquiring data on the changes in the internal grain structure, phase transformation, and defect evolution of mold materials after multiple injection molding cycles, and their impact on macroscopic properties (such as the coefficient of thermal expansion and elastic modulus), through experimental testing, historical data analysis, or multi-scale material simulation. The aim is to capture the performance degradation of mold materials caused by factors such as thermal fatigue and stress relaxation during long-term use. Obtaining viscoelastic deformation data of polymer materials during long-term service can be understood as acquiring characteristic data on the changes in the viscoelastic behavior (such as creep, stress relaxation, and shrinkage) of polymer materials over time under different environmental conditions such as temperature, stress, and humidity, through rheological testing, creep testing, and stress relaxation testing. The aim is to accurately describe the complex changes in the size and shape of polymer materials over time during curing and cooling processes and subsequent use.

[0040] In practical applications, data on the microstructural evolution of mold materials under long-term thermal cycling and the viscoelastic deformation data of polymer materials during long-term service are integrated with the deformation contribution of microscale deformation-sensitive points to the surface contour of the final product within a single injection molding cycle. This results in integrated data, which can be obtained by using data fusion algorithms or multiphysics coupling models to uniformly process and correlate data from different sources and time scales, constructing a comprehensive database or model. The purpose is to provide comprehensive input for subsequent accurate calculations and predictions. Furthermore, based on the integrated data, the changes in the local thermal expansion coefficient and elastic modulus of the mold material are calculated. This involves using the integrated data, through establishing a material constitutive model or employing machine learning algorithms, to quantify the specific changes in the thermal expansion coefficient and elastic modulus of the mold material in different regions and at different service stages. These changes reflect the impact of the microstructural evolution of the mold material on its macroscopic mechanical properties. Based on this, the changes in the local thermal expansion coefficient and elastic modulus of the mold material are introduced as correction terms into the calculation of the deformation contribution of microscale deformation-sensitive points to the final product surface contour within a single injection molding cycle. This aims to correct for the effects of time-varying mold material properties that may be ignored in traditional single-cycle simulations, making the calculation of deformation contribution more accurate. Simultaneously, based on the integrated data, the changing trend of local shrinkage of the polymer material is predicted. The prediction result refers to the prediction of the change in local shrinkage of the polymer material over time during curing and cooling, as well as long-term service, based on the viscoelastic deformation data of the polymer material and the geometric characteristics of the mold cavity, using finite element analysis or a specialized shrinkage prediction model. Finally, the correction terms and prediction results are superimposed to obtain the cumulative deformation contribution of microscale deformation-sensitive points to the final product surface contour throughout the entire product lifecycle. The superposition process can employ linear superposition, weighted averaging, or more complex nonlinear coupling models to comprehensively consider the long-term deformation effects of both the mold material and the polymer material.

[0041] This application addresses the limitations of considering only the deformation contribution of a single injection cycle by incorporating microstructural evolution data of the mold material under long-term thermal cycling and viscoelastic deformation data of the polymer material during long-term service. Specifically, the microstructural evolution of the mold material leads to changes in its local coefficient of thermal expansion and elastic modulus, which directly affect the actual size and shape of the mold cavity, and consequently the surface profile of the final product. By introducing these changes as correction terms into the deformation contribution calculation within a single injection cycle, the actual state of the mold during long-term use can be more accurately reflected. Simultaneously, the viscoelastic deformation characteristics of the polymer material determine its shrinkage behavior during curing and cooling, as well as during long-term service. By predicting the trend of local shrinkage changes in the polymer material, this time-dependent deformation can be captured. Finally, by superimposing the correction terms for the mold material with the predicted results for the polymer material, the cumulative deformation effects of both the mold and the polymer material throughout the entire product lifecycle can be comprehensively considered, resulting in a more comprehensive and accurate cumulative deformation contribution of microscale deformation-sensitive points to the surface profile of the final product.

[0042] In some preferred embodiments, it is assumed that a mold for producing high-precision aspherical optical lenses needs to be designed. First, accelerated thermal cycling experiments are conducted on the mold material (e.g., a high-strength steel), combined with microstructure analysis techniques (such as scanning electron microscopy and X-ray diffraction) to obtain data on microstructural evolution, including grain size changes and dislocation density increases, after tens of thousands of thermal cycles. The impact of these changes on the material's local coefficient of thermal expansion and elastic modulus is quantified. Simultaneously, long-term creep and stress relaxation tests are performed on the polymer material used (e.g., an optical-grade PMMA) to obtain its viscoelastic deformation curves under different temperature and stress conditions, thereby predicting the trend of local shrinkage during long-term service. Subsequently, this long-term data is integrated with the deformation contribution caused by microscale deformation-sensitive points within a single injection cycle, simulated using a local high-resolution solver. For example, in a specific region of the mold cavity, the local coefficient of thermal expansion of the mold material increases by 0.5% and the elastic modulus decreases by 2% due to long-term thermal cycling; these changes are calculated as correction terms. Meanwhile, it is predicted that the local shrinkage of the polymer material will increase by an additional 0.1% after five years of product service. These corrections and predictions are then superimposed, for example, through a weighted average model where the weight of long-term effects increases over time, to obtain the cumulative deformation contribution of this microscale deformation-sensitive point to the final lens surface profile over the lens's entire lifespan (e.g., ten years). For example, the final calculation shows that the cumulative deformation caused by this sensitive point may reach 50 nanometers, far exceeding the 10-nanometer deformation in a single cycle. Based on this cumulative deformation contribution, a more robust mold compensation strategy can be generated to ensure that the lens maintains its design accuracy throughout its entire lifespan.

[0043] This application further proposes a step to superimpose the correction term and the prediction result to obtain the cumulative deformation contribution of the microscale deformation sensitive point to the final product surface profile throughout the entire product life cycle, including: Receive correction terms and prediction results; Identify the nonlinear correlation between the correction term and the prediction result; Based on the nonlinear correlation, adjust the weights of the correction term and the prediction results at different time scales; The adjusted correction terms and prediction results are superimposed to obtain the cumulative deformation contribution of microscale deformation sensitive points to the final product surface profile throughout the entire product life cycle. The cumulative deformation contribution after stacking is verified to ensure that the cumulative deformation contribution meets the nanometer-level accuracy requirements.

[0044] Specifically, receiving correction terms and prediction results refers to the system acquiring correction terms formed by changes in the local thermal expansion coefficient and elastic modulus of the mold material, and prediction results formed by the trend of local shrinkage of the polymer material. These data form the basis for quantifying the cumulative deformation contribution. Identifying the nonlinear correlation between correction terms and prediction results can be understood as using advanced analytical techniques such as data mining and machine learning algorithms (e.g., neural networks, support vector machines, or Gaussian process regression) to analyze the interaction patterns of correction terms and prediction results under different operating conditions, different material batches, and different service times, to reveal potential nonlinear dependencies between them. The aim is to more accurately understand and model the complex coupling behavior of the mold and polymer material during long-term service. In practical applications, adjusting the weights of correction terms and prediction results at different time scales based on the nonlinear correlation involves dynamically assigning different influence factors to the correction terms and prediction results based on the identified nonlinear correlation model. For example, during short injection molding cycles, the influence of transient thermal expansion and elastic deformation may be emphasized; while during long-term service, the weights of viscoelastic creep of the polymer material and fatigue damage of the mold material may need to be increased. The aim is to ensure that the calculation of cumulative deformation contribution adapts to the dynamic evolution of material properties throughout the entire product lifecycle. Superimposing the adjusted correction terms and prediction results yields the cumulative deformation contribution of microscale deformation-sensitive points to the final product's surface profile throughout the product lifecycle. This involves comprehensively calculating the correction terms and prediction results, considering nonlinear correlations and time-scale weighting adjustments, to obtain a more accurate and comprehensive cumulative deformation contribution. Verifying the superimposed cumulative deformation contribution to ensure it meets nanometer-level accuracy requirements involves comparing it with actual measurement data (e.g., inspecting actual manufactured aspherical products using a high-precision optical profilometer) or cross-validating it using independent finite element analysis models to assess the accuracy of the calculated cumulative deformation contribution. The goal is to ensure that optimized designs achieve the extremely high precision standards required for aspherical optical components.

[0045] This application's solution addresses the problem of traditional simple superposition methods failing to accurately capture the coupling of complex material behaviors by introducing the identification and weight adjustment of nonlinear correlations between correction terms and prediction results. Because the microstructural evolution of mold materials and the viscoelastic deformation of polymer materials are not simple linear superposition relationships, but rather involve complex mechanisms of mutual influence and time-varying behavior, this solution, by identifying these nonlinear correlations, enables a deeper understanding of the material's true response during long-term service. Based on this, the weights at different time scales are adjusted according to the nonlinear correlations, allowing the calculation of cumulative deformation contribution to dynamically adapt to the decay of material properties and changes in environmental conditions, thus avoiding calculation biases caused by fixed weights or simple superposition. Finally, rigorous verification of the superimposed cumulative deformation contribution ensures the reliability of the entire optimization design process and meets the stringent nanometer-level precision requirements of aspherical products.

[0046] In some preferred embodiments, it is assumed that during injection molding of an aspherical mold, there is a nonlinear relationship between the change in the local thermal expansion coefficient of the mold material (correction term) and the trend of the local shrinkage of the polymer material (prediction result). For example, under high temperature and high pressure, the shrinkage behavior of the polymer is more sensitive to the microscopic deformation of the mold, while it exhibits different responses under low temperature and low pressure. Specifically, firstly, the system receives the correction term and prediction result calculated from the microstructure evolution data of the mold material and the viscoelastic deformation data of the polymer material. Next, a deep learning-based regression model is used to train historical injection molding data to identify the nonlinear correlation between the correction term and the prediction result. For example, the model may reveal that in the early stages of injection molding, the rapid cooling and shrinkage of the polymer contributes significantly to the final deformation, while in the later stages, the creep and fatigue accumulation effects of the mold material gradually become dominant. Based on the identified nonlinear correlation, the system dynamically adjusts the weights of the correction term and the prediction result at different time scales. For example, in the first few seconds of a simulated injection molding cycle, the predicted polymer shrinkage may be given higher weight; while when simulating thousands of cycles of mold operation over a long period, the weight of correction terms for changes in the mold material's thermal expansion coefficient and elastic modulus will increase accordingly. These weighted correction terms and the predicted results are then superimposed to obtain a more accurate cumulative deformation contribution of microscale deformation-sensitive points to the final product's surface profile throughout its entire lifecycle. Finally, to ensure that the cumulative deformation contribution meets nanometer-level accuracy requirements, a high-precision three-dimensional optical profilometer can be used to sample and inspect actual aspherical products, comparing their surface profile data with the simulated and predicted cumulative deformation contribution. If the deviation exceeds a preset nanometer-level threshold, the nonlinear correlation model or weight adjustment strategy needs to be iteratively optimized until verification is successful.

[0047] The steps for verifying the cumulative deformation contribution after stacking include: Continuously monitor the microstructure of mold materials and the viscoelastic response characteristics of polymer materials to obtain corresponding current performance data; Based on the deviation between the current performance data and the initial design parameters, calculate the current changes in the local thermal expansion coefficient and elastic modulus of the mold material, as well as the current trend of the local shrinkage of the polymer material. Based on the current changes in the local thermal expansion coefficient and elastic modulus of the mold material, and the current trend of the local shrinkage of the polymer material, the parameters used to verify the contribution of cumulative deformation are adjusted. The verification criteria for cumulative deformation contribution are redefined based on the current performance degradation of the mold material and polymer material. The cumulative deformation contribution after stacking is verified based on the adjusted verification parameters and verification criteria.

[0048] Specifically, continuous monitoring of the microstructure of the mold material and the viscoelastic response characteristics of the polymer material aims to acquire real-time performance data of the mold and polymer under actual operating or service conditions. This current performance data may include, but is not limited to, information on the internal structure of the material obtained through non-destructive testing techniques (such as ultrasound, X-ray diffraction, and infrared spectroscopy), and viscoelastic parameters obtained through mechanical tests (such as creep and stress relaxation tests). The goal is to capture the dynamic behavior of the material as a function of time, temperature, and pressure, providing real-time and accurate input for subsequent verification.

[0049] Specifically, calculating the current changes in the local thermal expansion coefficient and elastic modulus of the mold material, as well as the current trend of local shrinkage of the polymer material, based on the deviation between current performance data and initial design parameters, involves comparing real-time monitored material performance data with the ideal or initial parameters set during mold design. This comparison quantifies the actual changes in the local thermal expansion coefficient and elastic modulus of the mold material due to long-term thermal cycling, fatigue, and other factors, as well as the dynamic trend of local shrinkage behavior of the polymer material during curing and cooling. These changes and trends reflect the actual degradation or evolution of material properties and are key correction factors for accurate verification.

[0050] Furthermore, based on the current changes in the local thermal expansion coefficient and elastic modulus of the mold material, as well as the current trend of local shrinkage of the polymer material, the parameters used to verify the contribution of cumulative deformation are adjusted. This means that the verification model no longer relies on static, preset parameters, but is corrected in real time according to the actual dynamic changes of the material. For example, the material parameters in the finite element analysis model can be dynamically adjusted to better reflect the actual state of the mold and polymer, thereby improving the accuracy of the verification.

[0051] Simultaneously, the verification criteria for cumulative deformation contribution should be reset based on the current performance degradation of the mold and polymer materials. As the mold and polymer materials undergo prolonged use, their performance will gradually decline, and the original nanometer-level precision verification criteria may need to be dynamically adjusted to adapt to the actual state of the materials. For example, in the later stages of mold service, the precision requirements for certain local areas can be appropriately relaxed, or strict standards can be maintained for critical areas to achieve more reasonable verification and ensure that the verification criteria match the actual performance of the materials.

[0052] Finally, the cumulative deformation contribution after stacking is verified based on the adjusted verification parameters and verification criteria. This step uses updated model parameters and dynamically adjusted verification criteria to re-evaluate and confirm the cumulative deformation contribution of the previously calculated microscale deformation sensitive points to the final product surface profile throughout the entire product lifecycle, ensuring that it still meets the requirements of actual production and product performance under the current material conditions.

[0053] This application's solution addresses the problems of static material parameters and fixed verification standards in traditional verification methods by introducing a continuous monitoring and dynamic evaluation mechanism for mold and polymer materials. Specifically, the current performance data obtained through continuous monitoring can reflect the microstructural evolution and viscoelastic response characteristics of the materials in real time, thereby capturing the real changes that occur in the materials during long-term service. Based on this dynamic data, the current changes in the local thermal expansion coefficient and elastic modulus of the mold material, as well as the current trend of local shrinkage of the polymer material, can be accurately calculated. These quantitative indicators directly reflect the degradation or evolution of material performance. Therefore, the parameters used to verify the cumulative deformation contribution can be dynamically adjusted, enabling the verification model to more accurately simulate the deformation behavior under the current material state. Simultaneously, the verification standards are reset according to the actual degree of performance degradation of the materials, avoiding distortion of verification results due to material aging and ensuring the rationality and effectiveness of the verification process. This dynamic and adaptive verification mechanism allows the verification results of the cumulative deformation contribution to more realistically reflect the product's accuracy performance throughout its entire product lifecycle.

[0054] The steps for continuously monitoring the microstructure of mold materials and the viscoelastic response characteristics of polymer materials to obtain corresponding current performance data include: In specific areas of the mold material, micro-strain sensor arrays and micro-temperature sensor arrays are deployed to collect local strain data and local temperature data. The specific areas refer to those regions that have a significant impact on the accuracy of the final product, have complex geometric features, or have drastic curvature changes, as identified through geometric and thermodynamic analysis. Analyze local strain data to identify strain anomaly regions that exceed the normal thermal expansion range; Analyze local temperature data to identify temperature anomaly areas that deviate from the normal thermal cycling curve; Compare the strain anomaly region with the temperature anomaly region. If the two overlap in spatial location, the overlapping region is determined to be a local defect initiation region caused by abnormal stress concentration or material fatigue. If the abnormal strain region does not overlap with the abnormal temperature region, it is determined to be a microstructure evolution caused by normal thermal cycling.

[0055] Identifying "specific regions" is crucial, aiming to concentrate limited monitoring resources on areas most prone to deformation or defects, thereby improving monitoring efficiency and accuracy. These regions are typically areas within mold cavities with complex geometries, dramatic curvature changes, or those subjected to high stress and thermal gradients during injection molding. By deploying "micro-strain sensor arrays" and "micro-temperature sensor arrays," high-resolution, real-time or near-real-time acquisition of local strain and temperature in these critical areas can be achieved. The micro-sensor arrays provide detailed spatially distributed data, rather than data from a single measurement point, thus capturing localized changes at the microscale.

[0056] Furthermore, the collected local strain data is analyzed to identify "strain anomaly regions exceeding the normal thermal expansion range." The normal thermal expansion range can be preset using the known thermal expansion coefficient of the mold material and the actual operating temperature range, or obtained through benchmark testing. Any strain significantly deviating from this range may indicate mechanical effects other than thermal expansion, such as residual stress, plastic deformation, or crack initiation. Simultaneously, local temperature data is analyzed to identify "temperature anomaly regions deviating from the normal thermal cycling curve." The normal thermal cycling curve reflects the temperature change pattern of the mold under stable production conditions; any significant deviation from this curve, such as local overheating or abnormal heat dissipation, may indicate the presence of abnormal heat sources or changes in heat conduction paths within the mold.

[0057] As a preferred embodiment, this application emphasizes the importance of comparing "strain anomaly regions with temperature anomaly regions." When strain anomaly regions and temperature anomaly regions overlap spatially, it usually means that the region is simultaneously subjected to abnormal mechanical and thermal loads. This composite anomaly is a typical characteristic of "local defect initiation regions caused by abnormal stress concentration or material fatigue." For example, local overheating may lead to a decrease in material strength, making it more susceptible to abnormal strain under stress, ultimately accelerating the initiation of fatigue cracks. Conversely, if strain anomaly regions and temperature anomaly regions do not overlap, it indicates that these anomalies may be independent. For example, a simple strain anomaly may only be the microstructural evolution of the material under normal thermal cycling, while a simple temperature anomaly may only be a slight fluctuation in local heat dissipation efficiency; neither poses a serious risk of defect initiation.

[0058] This application's solution achieves multi-dimensional, high-precision monitoring of the microstructure of mold materials and the viscoelastic response characteristics of polymer materials by deploying micro-strain sensor arrays and micro-temperature sensor arrays in key areas of the mold material. It is precisely because of the simultaneous acquisition and collaborative analysis of local strain and temperature data that the system can effectively distinguish between microstructure evolution caused by normal thermal cycling and the initiation of local defects caused by abnormal stress concentration or material fatigue. Specifically, when strain anomalies and temperature anomalies overlap spatially, this composite signal provides strong evidence that there may be a risk of structural damage or defect initiation in that area, because abnormal stress concentration is often accompanied by local heating or abnormal heat conduction, and vice versa. This monitoring mechanism based on multi-physics coupling analysis significantly improves the early warning capability for potential defects and avoids misjudgments that may result from monitoring a single physical quantity.

[0059] In some preferred embodiments, assuming a high-precision aspherical optical lens is produced using an injection mold, preliminary geometric analysis and thermodynamic simulation identify specific regions as the edge areas and areas with the greatest curvature changes in the mold cavity. These regions significantly affect the final surface profile accuracy of the lens. Within these specific regions, miniature strain sensor arrays and miniature thermocouple arrays fabricated using MEMS technology are deployed. During the continuous injection molding process, these sensor arrays acquire local strain and temperature data in real time for this region.

[0060] The system first analyzes the collected local strain data. For example, by comparing it with the strain baseline of the mold material under standard thermal cycling, it finds that the strain value in a certain micro-region continuously exceeds the upper limit of the normal thermal expansion range, which is identified as a strain anomaly region. Simultaneously, the system analyzes the local temperature data. By comparing it with the normal thermal cycling curve of the mold under stable production conditions, it finds a region spatially similar to the aforementioned strain anomaly region, whose temperature curve shows a continuous local increase, which is identified as a temperature anomaly region.

[0061] The system then compares the spatial locations of the two abnormal areas. If these two abnormal areas overlap at the same minute location within the mold cavity—for example, at a sharp corner of the mold cavity where strain and temperature anomalies occur simultaneously—the system determines that the overlapping area is a region where local defects are initiating due to abnormal stress concentration or material fatigue. This determination triggers a higher-level alarm and may recommend further non-destructive testing or maintenance of the mold. Conversely, if only strain anomalies are detected in other areas without corresponding temperature anomalies, or only temperature anomalies are detected without corresponding strain anomalies, the system determines that these independent anomalies are microstructural evolutions caused by normal thermal cycling, such as slight creep or minor changes in the coefficient of thermal expansion of the material. These changes are within acceptable limits and do not require immediate intervention. In this way, the solution proposed in this application can accurately distinguish between normal aging of the mold and potential early defects, providing a precise basis for mold maintenance and parameter optimization.

[0062] This application further proposes steps for calculating the current changes in the local thermal expansion coefficient and elastic modulus of the mold material, including: Time-series sampling is performed on the current performance data of the mold material to obtain continuous measurement values ​​of the mold material; Trend analysis of continuous measurements of mold materials identifies long-term, slowly changing trend components, which reflect the inherent changes caused by the evolution of the mold material's microstructure. High-frequency components are extracted from continuous measurements of mold materials to identify rapid, non-periodic fluctuation components. These fluctuation components reflect transient disturbances caused by external environmental temperature fluctuations or measurement errors. The trend component is used as the current change in the local thermal expansion coefficient and elastic modulus of the mold material; High-frequency components are suppressed to eliminate the impact of transient disturbances on the calculation.

[0063] Specifically, time-series sampling of the current performance data of mold materials refers to continuously and periodically collecting various performance indicators of the mold materials within a certain time interval. For example, this involves continuously acquiring data such as strain and temperature through sensor arrays deployed in key areas of the mold, thus forming a series of continuous measurements arranged in chronological order. The purpose is to provide sufficient time-dimensional information for subsequent data analysis.

[0064] Trend analysis of continuous measurements of mold materials can be understood as using statistical or signal processing methods, such as moving averages, exponential smoothing, multinomial regression, or wavelet analysis, to separate components reflecting the long-term, slow changes in the mold material's performance from the continuous measurements. This trend component represents the inherent physical property changes caused by the evolution of the mold material's internal microstructure (such as grain size, grain boundaries, and dislocation density) under long-term thermal cycling and mechanical stress, resulting in slow shifts in its coefficient of thermal expansion or elastic modulus. The aim is to capture the essential, irreversible performance degradation or evolution of the material.

[0065] In practical applications, high-frequency component extraction from continuous measurements of mold materials refers to identifying and separating rapidly changing, non-periodic, or highly random fluctuation components from continuous measurements using techniques such as Fourier transform, wavelet decomposition, or high-pass filtering. These high-frequency components are typically caused by transient disturbances such as rapid changes in ambient temperature, instantaneous noise from measuring instruments, power supply fluctuations, or operational errors; they do not represent inherent changes in the mold material's properties. The aim is to separate these interference signals from the true performance data.

[0066] Furthermore, using the trend component as the current change in the local coefficient of thermal expansion and elastic modulus of the mold material means that after completing trend analysis and high-frequency component extraction, the identified trend component is directly used as an effective indicator to characterize the current changes in the coefficient of thermal expansion and elastic modulus of the mold material. This is because the trend component has eliminated the influence of transient disturbances and can more accurately reflect the true performance evolution of the material during long-term service.

[0067] Furthermore, high-frequency components are suppressed to eliminate the impact of transient disturbances on the calculation. This can be achieved using digital filtering techniques, such as low-pass filters, or by setting thresholds and removing outliers to remove or significantly attenuate the extracted high-frequency components from the data. The aim is to ensure that subsequent calculations and analyses are based solely on stable and reliable trend components, thereby improving the accuracy and stability of the calculation results.

[0068] This application's solution effectively addresses the impact of transient disturbances in mold material performance data on calculation accuracy by introducing time series analysis. Specifically, it first obtains continuous measurement values ​​containing rich temporal information by time-series sampling of the current performance data of the mold material. Based on this, trend analysis accurately identifies the long-term, slow-changing trend components reflecting the inherent changes caused by the microstructural evolution of the mold material. Simultaneously, high-frequency component extraction separates the rapid, non-periodic transient disturbance components caused by external environmental fluctuations or measurement errors. Subsequently, the purified trend components are used as the current changes in the local thermal expansion coefficient and elastic modulus of the mold material, thereby suppressing high-frequency components and ensuring the purity and accuracy of the calculation results. This approach ensures that the calculated changes truly reflect the intrinsic performance degradation of the mold material, rather than superficial fluctuations influenced by external interference.

[0069] In some preferred embodiments, it is assumed that a micro-strain sensor array and a micro-temperature sensor array are deployed in specific areas of the mold cavity to continuously monitor the strain and temperature data of the mold material. These sensors acquire data once per second, forming a continuous time series. To calculate the current changes in the local coefficient of thermal expansion and modulus of elasticity of the mold material, these continuous measurements are first sampled as a time series. For example, data from the past 24 hours can be processed every hour.

[0070] Specifically, after obtaining continuous measurements, trend analysis can be performed on the data using moving averages or Savitzky-Golay filters to identify long-term, slowly changing trend components. For example, by calculating a 24-hour moving average, the effects of intraday temperature fluctuations or short-term stress changes can be effectively smoothed out, thereby revealing the slow drift in strain or temperature caused by the microstructural evolution of the mold material due to long-term thermal cycling.

[0071] Meanwhile, by subtracting trend components from the raw data or using a high-pass filter, fast, non-periodic fluctuations can be extracted. These fluctuations may originate from the instantaneous start-up of the workshop's air conditioning system, minute changes in coolant flow rate, or even random noise from the sensor itself. For example, a high-pass filter can filter out signals with frequencies below a certain threshold (such as those changing once per hour), retaining only higher-frequency fluctuations.

[0072] Ultimately, the trend components obtained through trend analysis are used as the current changes in the local coefficient of thermal expansion and modulus of elasticity of the mold material. For example, if the trend components show a slow increase of 0.01% in the average strain of the mold material over the past month, this is considered an inherent change in its modulus of elasticity or coefficient of thermal expansion. High-frequency fluctuations identified are suppressed or ignored to ensure that these transient disturbances do not incorrectly affect the assessment of long-term material performance changes. In this way, more accurate and stable data on mold material performance changes can be obtained, providing a reliable basis for subsequent mold parameter optimization.

[0073] The steps for calculating the current trend of local shrinkage in polymer materials include: Time-series sampling is performed on the current performance data of polymer materials to obtain continuous measurement values ​​of polymer materials; Based on continuous measurements of polymer materials, the nonlinear viscoelastic response characteristics caused by the evolution of the internal microstructure of polymer materials are identified. Based on the nonlinear viscoelastic response characteristics, the prediction parameters used to predict the local shrinkage trend of polymer materials are dynamically adjusted. Based on the adjusted prediction parameters, the current trend of local shrinkage of polymer materials is predicted.

[0074] Time-series sampling of current performance data for polymer materials refers to continuously or periodically measuring the physical or chemical properties of polymer materials at different time points to obtain a series of data points arranged in chronological order. These data points may include, but are not limited to, dimensional changes, density, hardness, and molecular weight distribution. The purpose is to capture the dynamic behavior and performance degradation process of polymer materials under actual service conditions.

[0075] Furthermore, based on continuous measurements of polymer materials, identifying the nonlinear viscoelastic response characteristics caused by the evolution of the internal microstructure of polymer materials can be understood as using advanced data analysis on the sampled time-series data, such as machine learning algorithms, nonlinear regression models, or signal processing techniques, to reveal how changes in the internal molecular chain structure, crystallinity, crosslinking density, and other microstructures of polymer materials affect their macroscopic viscoelastic behavior during long-term use. Nonlinear viscoelastic response characteristics refer to the fact that when a material is subjected to stress or strain, its response is no longer a simple linear relationship, but exhibits complex nonlinear behavior related to factors such as time, loading history, and temperature. Identifying these characteristics is crucial for accurately predicting the long-term performance of materials.

[0076] Based on this, and according to the identified nonlinear viscoelastic response characteristics, the prediction parameters used to predict the local shrinkage trend of polymer materials are dynamically adjusted. This means that key parameters in the prediction model are corrected in real time or periodically based on the actual nonlinear viscoelastic behavior exhibited by the polymer material. For example, parameters such as the material's creep modulus, relaxation time spectrum, and glass transition temperature can be adjusted to enable the prediction model to more accurately reflect the material's current state and future shrinkage trend. This dynamic adjustment mechanism ensures the accuracy and adaptability of the prediction.

[0077] Finally, based on the adjusted prediction parameters, the current trend of local shrinkage of the polymer material is predicted. This involves using a dynamically adjusted prediction model, combined with current performance data and historical data, to quantitatively predict the trend of local shrinkage of the polymer material over a future period. This prediction result will provide crucial input for optimizing mold parameters to effectively compensate for the shrinkage effect of the polymer material, thereby ensuring the dimensional accuracy of the final product.

[0078] This application's solution, through refined time-series sampling of the current performance data of polymer materials, comprehensively captures the dynamic changes of materials under actual service conditions. By further identifying the nonlinear viscoelastic response characteristics caused by microstructural evolution, this solution can gain a deeper understanding of the intrinsic mechanism of material performance degradation, rather than merely remaining at the macroscopic level. It is precisely this accurate identification of nonlinear viscoelastic response characteristics that allows the parameters in the prediction model to be dynamically and in real-time adjusted, thereby overcoming the limitations of traditional fixed-parameter models in predicting the long-term shrinkage behavior of polymer materials. Through this dynamic adjustment mechanism, this solution can more accurately predict the current trend of local shrinkage of polymer materials, providing more reliable and accurate data support for subsequent mold parameter optimization.

[0079] refer to Figure 2 This application proposes a parameter optimization design system for aspherical molds based on dynamic modeling. This system aims to efficiently and accurately implement the aforementioned optimization design method. The system includes: The acquisition module is used to acquire the geometric data of the mold cavity; The identification module is used to analyze geometric data to identify areas that have a critical impact on the accuracy of the final product, as well as areas that do not have a critical impact. The critical region simulation module is used to allocate computing resources for regions with critical impacts and to perform simulations that reflect microscopic physical processes. The non-critical area simulation module is used to perform simulations for areas with non-critical impact, with computational efficiency as the primary consideration. The information exchange module is used to exchange and correlate physical field information between the simulation of critically affected areas and the simulation of non-critically affected areas to ensure the continuity of the overall simulation process. The physical field information includes data on temperature field, pressure field, flow velocity field and stress field.

[0080] Specifically, the acquisition module can be configured to receive mold cavity geometric model data from CAD software, 3D scanning equipment, or other data sources, and convert it into a unified data format that the system can process. Its purpose is to provide foundational data for subsequent analysis and simulation. The identification module can be understood as automatically or semi-automatically dividing the mold cavity into different regions using algorithms such as geometric feature analysis, curvature analysis, and wall thickness variation analysis. For example, high-curvature regions, thin-walled regions, and complex structural regions are typically identified as critical influence regions, while flat, regular regions are identified as non-critical influence regions. This aims to provide a basis for subsequent differentiated simulations. In practical applications, the critical region simulation module utilizes high-performance computing resources, such as GPU clusters or distributed computing systems, to perform high-precision finite element analysis or computational fluid dynamics simulations on critical influence regions. This captures the microscopic physical phenomena during melt flow, heat transfer, and solidification processes, ensuring that regions crucial to product accuracy receive the most detailed analysis. The non-critical region simulation module can be understood as employing simplified models, fast algorithms, or lower mesh densities for simulation. For example, it may use one-dimensional or two-dimensional models, empirical formulas, or fast iterative solvers. Its purpose is to ensure overall simulation efficiency while avoiding unnecessary computational overhead. The information exchange module can be configured to use boundary condition transfer, data interpolation, or coupling algorithms to provide the detailed results of the critical region simulation as input to the non-critical region simulation, or to feed back the macroscopic results of the non-critical region simulation to the critical region simulation. This achieves seamless data integration and a unified description of the physical processes between different simulation regions, ensuring that the simulation results for the entire mold cavity maintain physical consistency and coherence.

[0081] This application's system effectively solves the efficiency and coordination problems that traditional methods may face during execution by modularizing and integrating the above-mentioned method steps. The acquisition module ensures the accurate import of raw data, laying the foundation for subsequent analysis. The identification module intelligently divides the mold cavity, allowing for targeted allocation of computing resources and avoiding the enormous computational burden caused by performing equally high-precision simulations on all regions. The critical region simulation module focuses on performing in-depth and detailed microscopic physical process simulations on the regions with the greatest impact on accuracy, thereby ensuring the accurate capture of key details. The non-critical region simulation module improves computational efficiency by simplifying the model while ensuring the overall simulation coherence. Finally, the information exchange module acts as a bridge, ensuring the seamless transmission and correlation of physical field information between simulation regions of different precision, enabling the simulation results of the entire mold cavity to maintain a high degree of physical consistency and coherence, thereby avoiding data isolation or physical mismatch problems that may be caused by region division.

[0082] Through the above technical solutions, the system of this application can achieve automated, efficient, and integrated execution of the parameter optimization design method for aspherical molds based on dynamic modeling. The system significantly improves simulation efficiency through intelligent module partitioning and resource allocation, while ensuring simulation accuracy for key areas. Furthermore, the introduction of an information exchange module ensures seamless connection of physical field information between different simulation regions, avoiding data gaps or physical inconsistencies that may occur in traditional methods, thereby improving the reliability and accuracy of the overall simulation results. Therefore, this system provides a stable, efficient, and easy-to-operate platform for parameter optimization design of aspherical molds, greatly shortening the product development cycle and improving the accuracy and performance of the final product.

[0083] The content disclosed above is only a preferred and feasible embodiment of the present invention, and is not intended to limit the scope of protection of the present invention. Therefore, all equivalent technical changes made based on the content of the present invention specification and drawings are included within the scope of protection of the present invention. Furthermore, the elements therein can be updated as technology develops.

Claims

1. A method for optimizing the parameters of an aspherical mold based on dynamic modeling, characterized in that, The method includes the following steps: Obtain the geometric data of the mold cavity; Analyze the geometric data to identify areas that have a critical impact on the accuracy of the final product, as well as areas that have a non-critical impact. For regions with critical impacts, computing resources are allocated, and simulations that reflect microscopic physical processes are conducted. For areas with non-critical impact, simulations are conducted with computational efficiency as the primary consideration. The simulation of critically affected areas is conducted by exchanging and correlating physical field information between the simulation of non-critically affected areas to ensure the coherence of the overall simulation process. The physical field information includes data on temperature field, pressure field, flow velocity field, and stress field.

2. The method for optimizing the parameters of an aspherical mold based on dynamic modeling as described in claim 1, characterized in that, For regions with critical impacts, the steps of allocating computational resources and conducting simulations that reflect microscopic physical processes include: In the critical areas of the mold cavity, deploy a micro-sensor array to collect local temperature, pressure, and flow rate data; Analyze local temperature, pressure and flow rate data to identify transient microscale deformation sensitive points. These microscale deformation sensitive points refer to points in the mold cavity that cause deformation due to micro-fluid dynamics, local thermal gradients or changes during the curing process. For microscale deformation-sensitive points, a local high-resolution solver is activated to simulate the local flow, transient heat transfer, and polymer solidification process of the melt under the influence of microscale eddies or cavitation, and the simulation results are obtained. Based on the simulation results, the deformation contribution of microscale deformation-sensitive points to the final product surface profile is quantified. A compensation strategy is generated based on the deformation contribution.

3. The method for optimizing the parameters of an aspherical mold based on dynamic modeling as described in claim 2, characterized in that, The steps for analyzing local temperature, pressure, and flow velocity data to identify transiently emerging microscale deformation-sensitive points include: Local temperature, pressure, and flow rate data are decomposed to separate signal components caused by different microscale physical phenomena; Based on the decomposed signal components, identify transient microscale deformation sensitive points.

4. The method for optimizing the parameters of an aspherical mold based on dynamic modeling as described in claim 2, characterized in that, Based on the simulation results, the steps to quantify the deformation contribution of microscale deformation-sensitive points to the final product surface profile include: To obtain data on the microstructure evolution of mold materials under long-term thermal cycling; To obtain viscoelastic deformation data of polymer materials during long-term service; The data on the microstructure evolution of mold materials under long-term thermal cycling and the viscoelastic deformation data of polymer materials during long-term service are integrated with the deformation contribution of microscale deformation-sensitive points to the surface profile of the final product within a single injection cycle to obtain integrated data. Based on the integrated data, calculate the changes in the local thermal expansion coefficient and elastic modulus of the mold material; The changes in the local thermal expansion coefficient and elastic modulus of the mold material are introduced as correction terms into the calculation of the deformation contribution of microscale deformation-sensitive points to the surface profile of the final product within a single injection cycle. Based on the integrated data, the changing trend of local shrinkage of polymer materials is predicted, and the prediction results are obtained. By superimposing the correction terms and prediction results, the cumulative deformation contribution of microscale deformation-sensitive points to the final product surface profile throughout the entire product lifecycle is obtained.

5. The method for optimizing the parameters of an aspherical mold based on dynamic modeling as described in claim 4, characterized in that, The steps to overlay the correction terms and prediction results to obtain the cumulative deformation contribution of microscale deformation-sensitive points to the final product surface profile throughout the entire product lifecycle include: Receive correction terms and prediction results; Identify the nonlinear correlation between the correction term and the prediction result; Based on the nonlinear correlation, adjust the weights of the correction term and the prediction results at different time scales; The adjusted correction terms and prediction results are superimposed to obtain the cumulative deformation contribution of microscale deformation sensitive points to the final product surface profile throughout the entire product life cycle. The cumulative deformation contribution after stacking is verified to ensure that the cumulative deformation contribution meets the nanometer-level precision requirements.

6. The method for optimizing the parameters of an aspherical mold based on dynamic modeling as described in claim 5, characterized in that, The steps for verifying the cumulative deformation contribution after stacking include: Continuously monitor the microstructure of mold materials and the viscoelastic response characteristics of polymer materials to obtain corresponding current performance data; Based on the deviation between the current performance data and the initial design parameters, calculate the current changes in the local thermal expansion coefficient and elastic modulus of the mold material, as well as the current trend of the local shrinkage of the polymer material. Based on the current changes in the local thermal expansion coefficient and elastic modulus of the mold material, and the current trend of the local shrinkage of the polymer material, the parameters used to verify the contribution of cumulative deformation are adjusted. The verification criteria for cumulative deformation contribution are redefined based on the current performance degradation of the mold material and polymer material. The cumulative deformation contribution after stacking is verified based on the adjusted verification parameters and verification criteria.

7. The method for optimizing the parameters of an aspherical mold based on dynamic modeling as described in claim 6, characterized in that, The steps for continuously monitoring the microstructure of mold materials and the viscoelastic response characteristics of polymer materials to obtain corresponding current performance data include: In specific areas of the mold material, micro-strain sensor arrays and micro-temperature sensor arrays are deployed to collect local strain data and local temperature data. The specific areas refer to those regions that have a significant impact on the accuracy of the final product, have complex geometric features, or have drastic curvature changes, as identified through geometric and thermodynamic analysis. Analyze local strain data to identify strain anomaly regions that exceed the normal thermal expansion range; Analyze local temperature data to identify temperature anomaly areas that deviate from the normal thermal cycling curve; Compare the strain anomaly region with the temperature anomaly region. If the two overlap in spatial location, the overlapping region is determined to be a local defect initiation region caused by abnormal stress concentration or material fatigue. If the abnormal strain region does not overlap with the abnormal temperature region, it is determined to be a microstructure evolution caused by normal thermal cycling.

8. The method for optimizing the parameters of an aspherical mold based on dynamic modeling as described in claim 6, characterized in that, The steps for calculating the current changes in the local coefficient of thermal expansion and the modulus of elasticity of the mold material include: Time-series sampling is performed on the current performance data of the mold material to obtain continuous measurement values ​​of the mold material; Trend analysis of continuous measurements of mold materials identifies long-term, slowly changing trend components, which reflect the inherent changes caused by the evolution of the mold material's microstructure. High-frequency components are extracted from continuous measurements of mold materials to identify rapid, non-periodic fluctuation components. These fluctuation components reflect transient disturbances caused by external environmental temperature fluctuations or measurement errors. The trend component is used as the current change in the local thermal expansion coefficient and elastic modulus of the mold material; High-frequency components are suppressed to eliminate the impact of transient disturbances on the calculation.

9. The method for optimizing the parameters of an aspherical mold based on dynamic modeling as described in claim 6, characterized in that, The steps for calculating the current trend of local shrinkage in polymer materials include: Time-series sampling is performed on the current performance data of polymer materials to obtain continuous measurement values ​​of polymer materials; Based on continuous measurements of polymer materials, the nonlinear viscoelastic response characteristics caused by the evolution of the internal microstructure of polymer materials are identified. Based on the nonlinear viscoelastic response characteristics, the prediction parameters used to predict the local shrinkage trend of polymer materials are dynamically adjusted. Based on the adjusted prediction parameters, the current trend of local shrinkage of polymer materials is predicted.

10. A parameter optimization design system for aspherical molds based on dynamic modeling, applying the parameter optimization design method for aspherical molds based on dynamic modeling as described in claim 1, characterized in that, The system includes: The module acquires the geometric data of the mold cavity; The identification module analyzes geometric data to identify areas that have a critical impact on the accuracy of the final product, as well as areas that do not have a critical impact. The critical region simulation module allocates computing resources to key affected areas and performs simulations that reflect microscopic physical processes. The non-critical area simulation module performs simulations for areas with non-critical impact, with computational efficiency as the primary consideration. The information exchange module facilitates the exchange and correlation of physical field information between the simulation of critically affected areas and the simulation of non-critically affected areas, ensuring the continuity of the overall simulation process. The physical field information includes data on temperature, pressure, flow velocity, and stress fields.

Images